_pickle.UnpicklingError: invalid load key, '\x0a'

[yang1997yi]

hello，I meet this error：
Traceback (most recent call last):
File "SuperGAT/main.py", line 471, in
many_seeds_result = run_with_many_seeds(main_args, num_total_runs, gpu_id=alloc_gpu[0])
File "SuperGAT/main.py", line 403, in run_with_many_seeds
ret = run(_args, gpu_id=gpu_id, **kwargs)
File "SuperGAT/main.py", line 286, in run
**dataset_kwargs,
File "/data1/yangyi/SuperGAT/SuperGAT/data.py", line 634, in get_dataset_or_loader
dataset = dataset_cls(root=root, **kwargs)
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/datasets/planetoid.py", line 33, in init
super(Planetoid, self).init(root, transform, pre_transform)
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/data/in_memory_dataset.py", line 53, in init
pre_filter)
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/data/dataset.py", line 93, in init
self._process()
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/data/dataset.py", line 166, in _process
self.process()
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/datasets/planetoid.py", line 58, in process
data = read_planetoid_data(self.raw_dir, self.name)
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/io/planetoid.py", line 19, in read_planetoid_data
items = [read_file(folder, prefix, name) for name in names]
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/io/planetoid.py", line 19, in
items = [read_file(folder, prefix, name) for name in names]
File "/data1/yangyi/anaconda3/envs/pt1.4/lib/python3.6/site-packages/torch_geometric/io/planetoid.py", line 66, in read_file
out = pickle.load(f, encoding='latin1')
_pickle.UnpicklingError: invalid load key, '\x0a'.
Do you know how to solve it?my python version is 3.6，Thanks a lot。

[dongkwan-kim]

Hi, it looks like an error raised from torch_geometric's internal codes.
I have never seen this kind of error, so I am not sure what the cause exactly is.
Instead, I will let you know my system information.

• docker image: nvidia/cuda:10.0-cudnn7-devel-ubuntu16.04
• python: 3.7.6